MattB 34 Posted March 24 Posted March 24 Hi all, New member here, but I've owned this stack of Carver gear for a little over 20 years! I originally bought a M1.0t and C-1 from a local repair shop that repaired them and the customer never paid for the repairs. Picked em both up for $100. What a steal! Not long after, I added an M1.5t to the list. They spent several years in various configurations of speakers, subs, 2-channel and surround duty. At some point, I picked up a pair of JBL L7 and an Acurus A250 amp Add in life events, house moves, and other things, and these toys got put on a shelf in a closet for about 10 years. I recently decided it was time to clean out the closets and sell off the stuff I wasn't using, so I dusted off these amps and plugged them in to check operation. POP! went the fuses on both! surprise- the caps failed while sitting on the shelf. As an electrical engineer and fan of DIY, I decided to attempt to repair these things myself-- BEFORE learning about this site and Nelion I started digging around and found recap kits on ebay, so I ordered what I could for both the M1.0t and M1.5t. After doing a fair amount of reading, I've learned I bought parts from a seller with a poor reputation, so I'm sorry to the guys who's efforts were stolen- I screwed up out of ignorance! I started on the M1.0t since the single large board looked easier to access all of the terminals, and the service manual was easier to read! Getting the old multi-caps out of was a bit of a pain, but I was able to replace them and recap the rest of the board in a couple evenings and recheck the bias. The M1.0t was relatively easy to repair and it sounds wonderful again! They drive these big JBLs with a nice punch, and the warm sound is much nicer than the Acurus which seems a bit sterile in comparison. Last week, I started in on the M1.5t, and it needed a lot more help. Previous repairs were sketchy with different size caps on both sides of the 50/80V rails and just glued on top of the old failed capacitors! It also appeared to have been owned by a smoker prior to me as the inside was covered in something sticky and then dust accumulated on it. I decided a full cleaning would be appropriate for this one, and removing the amp cards from the chassis would make it easier to get to everything for cleaning and access the capacitors for replacement. NOW for the tech question! I've gotten all of that done and followed the service manual to set voltage and bias (122V at the +120V rail, and 4mV across the big resistor).. unfortunately, the amp is idling a bit hot- it's drawing ~1.1A at idle and a fair amount of heat is originating from the center of the chassis looks like Q5 and Q6 on the right channel when I sweep my thermal camera across the chassis. Also, the two large 1/2watt resistors on the lower edge of the power supply board are quite hot- the board around them was already discolored when I opened the chassis, so I removed them from the board and checked resistance- they meet published spec (3.8k and 6k I think?). I don't have a ton of tools perform a full load test and measure distortion, but I was able to drive the thing with 1k sine wave into some 5ohm power resistors to about 20V P-P. The waveform looked nice and clean. With that, I hooked it up to my speakers and ran some low volume music through them. everything sounds great, maybe a bit more "dry" than the M1.0t, but the clarity and imaging is all there. so maybe its characteristic sound is just a bit more dry? I let it run for quite a while at low volume and the entire chassis is warm to the touch- maybe 120F. This has me concerned that the high idle current is a problem. Do I simply adjust the bias voltage and current a bit lower- like 1-2mV, or do I do some more digging? unfortunately my cheap $20 oscilloscope only has 50V max voltage, so I can't poke around too much with it. I'd love any advice you guys can provide to try to finish this project up. Trying to figure out the layout of this forum-- will upload some pics when I can figure it out. (I no longer have a web host to upload and link files, but only see a way to link photos here) 8
Kurt 2,483 Posted March 25 Posted March 25 I can't answer the electrical questions. But, I have an m1.0t and several m1.5t's. All have been refurbished/upgraded by Nelion to mkII status. None of them get hot at all in any place... they all run fairly cool even if driven hard (I've measured them running hard at warmest spot on the chassis and I get about 95 deg F). I have fans, but I don't use them. As long as you provide enough space around the amps, they stay nice and cool. The m1.5t MkII's I have are different sounding than the m1.0t MkII......But, by just a bit. Not really all that different. If I had to characterize it, I'd say the m1.5t's are "brighter"...ie a little more high end. Kind of like the difference between CD's and Records, when CD's first came out... The m1.0t mkII is a little "warmer" sounding. When I pair my m1.5t's with speakers that have a lot of high end, it can feel kinda harsh on certain songs. But, really, like I said, this is nit picking differences. I've run them back to back on same speakers in blind tests, and I really can't tell them apart that easily. (Note: I had Nelion upgrade the m1.0t to MkII Opt 2, which gives it almost as much power as the m1.5t mkII refurbishment). Hope you figure out your issue.... lots of help here on the Carver Site. Oh, and welcome aboard! 1
Rob 5,267 Posted March 25 Posted March 25 Welcome to the site. Do you have a 1x/10x probe for your scope?
timmyjim 363 Posted March 25 Posted March 25 Properly adjusted, the m1.5t does idle a bit higher than most of the other amps, but it should be under 500mA as is specified in most service manuals. Should be in the 400-450mA ballpark. The service manual indicates that you should adjust bias across R87 to 4mV as you have done. I have not worked on a m1.5t in a while, but I don't believe this is accurate. I always adjusted bias across BOTH emitter resistors, in fact always connected my meter to the emitters, not the resistors. Easier. As a sanity check, I reviewed the pm1.5 manual to see if it suggests the same bias adjust. In fact it indicates as I have mentioned above, connecting the meter across BOTH R49 and R87. It also suggests a level of 3.4mV. This is the method you should use. I think your amp will be fine and indicate less than 500mA of idle current when adjusted this way. I might suggest that you use the pm1.5 schematics for any future amp card debug. They are easier to read, and match about 99%. 1
BobTFM35 2,571 Posted March 25 Posted March 25 Welcome to the Carver Site! Nice repair on the 1.0. I can only imagine what the look on your face was seeing the inside of the 1.5 and the glued caps. I just hope that is was not a trained professional doing that sort of repair. As for attaching images I usually edit mine in a separate photo editor to reduce the size a bit. After that I copy and paste the image into my post. Here is an example of the result of copy and paste of the image from a cap replacement I did on a cassette deck. Size shown is about 128 kb. It will then be made part of your "My Attachments" for future use in other posts. And I did cut them off. Good luck on finishing off the 1.5, and looking forward to seeing pictures of this and future projects. 1
Community Admin AndrewJohn 9,914 Posted March 25 Community Admin Posted March 25 Welcome to TCS, @MattB, glad you found us. Another resource is to look through the amp repair threads in the amp forum (scroll down on the home page). Several threads there may be useful. FWIW, I have three m1.5t amps, and they do not run hot at all. When you say "sounds dry", I relate that to sounding crisper, sharper, and with more piercing "highs". That's typical. But if "dry" to you means lifeless, I'd keep looking. This amp is definitely not lifeless. I love it! Again, welcome and glad you are here. 1
MattB 34 Posted March 25 Author Posted March 25 (edited) Hi all, many thanks for the welcome and the pointers. There have been several replies, so I'll try to do it without making a 14 page reply. - Will revisit the biasing based on @timmyjim comments. I believe you are right on checking bias across both emitters. When I first powered up the amp, the bias was at about 1-2mV across R87 on both channels. Doubling that reading to account for both emitters would put me in the 3-4mV range which matches the PM1.5 and your recommendation. Given how much handling I had done of the amp cards while they were out of the chassis, I assumed I bumped the pots on both. I blindly did what the manual said believing whoever wrote the manual was the expert! I downloaded the PM1.5 manual and diagrams off the manual repository last night. the Member Diagram pdf is awesome! And yes clipping to the emitter pins is way easier than grabbing that lower leg of R49/R87. I clipped my lead to the Q5 emitter and the top leg of the resistor. I wasn't the first in there either as the insulation on the resistor leg had been picked away for access. I also do not remember this amp running warm in the past, but again it's been sitting for easily 10 years- I considered it a win the thing didn't blow a breaker or catch fire when I powered it up! - @Rob, unfortunately no 10x probe on hand, but I'm ordering one today. My oscilloscope is a little handheld $20 junklet I bought on Amazon to commission some outdoor siren arrays for work, and it only came with a BNC to alligator clip connector. It worked for what I needed at the time. - As for the amp sounding "dry", I don't insinuate that it sounds weak, bad, or lifeless at all, it just doesn't have the warmth that the M1.0t has. Maybe the highs are a bit exaggerated compared to the M1.0t? I didn't listen to it very long or do a real A/B test with same material, I was just plugging in cables and ran it through a couple songs. I did replace C35 - the 0.33uF cap right at each channel's inputs. as the signal goes through it before anything else, I expect that cap to be essential to the sound. no idea what brand it was- it's a mylar or polypropylene ~250V cap I found in my surplus pile after breaking one of the leads off the original 'red' ones glued to the board and corroding around the legs. And now for a few pics- I didn't know you could just drag and drop photos in line - I haven't done forum stuff in years! Before: You can see how dirty the inside was and all the sticky buildup on the traces- hard to get clean measurements and good solder joints when everything is coated and smells like an ashtray when you heat it up! You can also see the random 2200uF cap glued to the original 50/80 caps- there's a 6800uF on the other side! This poor unbalanced amplifier! In the middle of the cleanup and reassembly work: (I replaced the mica insulators on the bottom with some kapton tape and cleaned/ replaced all of the thermal paste while testing the output transistors. No more globs of white messy goo! And back together: Another minor mod was to replace the 'lamp cord' power cable with a fused C14 socket and power switch. Back together and running for a little bit! Rest of the System specs: Pioneer AVR 7.1 receiver to run it all (forget the model #, but it's irrelevant) Sony SACD player (yeah, I bought into it many years ago!) M1.0t and M1.5t to run the JBL L7s in a biamp configuration. SurgeX power sequencer Velodyne HGS-18 sub. A freebie from my neighbor cause it had an annoying 60hz hum. replaced the caps in the audio circuit and voila! Some sort of in-wall JBLs for center and surround duties, driven off the receiver's internal amp. Next upgrade: beefier 120VAC mains cabling with multiple circuits since this whole mess is capable of about roughly 3000W RMS / 5000w dynamic power! now will I ever use it in a 16x20ft room? probably not more than a few seconds.... Many appreciates for the warm welcome and advice! Edited March 25 by MattB error/formatting cleanup 6
BobTFM35 2,571 Posted March 25 Posted March 25 Great pictures and system!! And now you are a seasoned pro!!
timmyjim 363 Posted March 25 Posted March 25 On 3/24/2025 at 5:45 PM, MattB said: a fair amount of heat is originating from the center of the chassis looks like Q5 and Q6 on the right channel when I sweep my thermal camera across the chassis. Also, the two large 1/2watt resistors on the lower edge of the power supply board are quite hot- the board around them was already discolored when I opened the chassis A lot of Carver amps have hot spots. Fairly normal. The m1.5t does get hot right in the middle. .... iirc the intersection of the back amp card and the main board. On 3/24/2025 at 5:45 PM, MattB said: removing the amp cards from the chassis Impressive.
MattB 34 Posted March 26 Author Posted March 26 Wife and kids were out of the house tonight so I had a chance to set up shop and go through biasing again. across both emitters, I was seeing almost 9mV and idle current of 1.8A. Lots of wasted heat! After turning bias back down to 3.4-3.5mV, idle current is now ~850mA. No further improvement if I reduced to minimum, so I guess that's that? I cobbled together a little more stuff and did a short test of 1kHz full power into some 5ohm power resistors. 54.5V RMS into 5ohms = 594W. There's a bit more room in there but I couldn't see the waveform well on my little scope so I couldn't tell where the clipping began and I didn't want to leave it at full power for long- these chassis just don't have much of a heat sink and was heating up quickly! I put the cover back on and let it cool down while mounting it back in the rack. After playing some low level signals for a while, the faceplate of the chassis is now idling at ~82F while my M1.0t is at 79F just below it. This makes me feel MUCH better as previously it was idling at >120F. I also took a few thermal images for fun while I had the cover off. It's interesting which little resistor bundles on each board are toasty warm while others have nothing. Will post those tomorrow once I transfer off my phone. 3
timmyjim 363 Posted March 26 Posted March 26 7 hours ago, MattB said: idle current is now ~850mA. Not sure how you are measuring this? Assuming an accurate measurement, it's still too high. 7 hours ago, MattB said: I cobbled together a little more stuff and did a short test of 1kHz full power into some 5ohm power resistors. 54.5V RMS into 5ohms = 594W. Unless you have disabled some of the "speaker protection" circuitry, this must have been a very short test? Carver added circuitry to the m1.5t that seems more marketing driven than anything else. It will put the amp into protection after a short time at rated power. This circuitry can be disabled by lifting 2 diodes on the main board. You can figure this out easily by comparing the pm1.5 and m1.5t main board schematics. With these diodes removed, the m1.5t becomes a 450WPC@8ohm amplifier .... no other modification required. 1
MattB 34 Posted March 26 Author Posted March 26 1 hour ago, timmyjim said: Not sure how you are measuring this? Assuming an accurate measurement, it's still too high. Measurement is using a Klein clamp-on style amp meter around the 120VAC line wire. Bias is measured concurrently using a Fluke multimeter. Its no NIST calibrated unit, but I would think it's close enough for gubmint work- not off by factor of 2! ....Aaaaaand I just did a little digging online and this model doesn't seem to be well regarded on the accuracy front. Sooo, now I'm doubting the measurements just as you are! Maybe I'll hook the Fluke up in-line and take a proper measurement. That will also tell me whether I can trust the Klein or not in the future... At least the good news is that with this bias setting, the transformer only has minimal hum and doesn't even get warm to the touch at idle, whereas it stayed quite warm before. 1 hour ago, timmyjim said: Unless you have disabled some of the "speaker protection" circuitry, this must have been a very short test? Carver added circuitry to the m1.5t that seems more marketing driven than anything else. It will put the amp into protection after a short time at rated power. This circuitry can be disabled by lifting 2 diodes on the main board. You can figure this out easily by comparing the pm1.5 and m1.5t main board schematics. With these diodes removed, the m1.5t becomes a 450WPC@8ohm amplifier .... no other modification required. Yes, these tests were only for a short time- about 15-20sec since these things don't have a massive heat sink and aren't intended for long term brute force. Bob did a good job of reducing materials to just what was necessary to make music, which is very peaky by nature. I did turn it up once with no load until clipping and the protection circuit kicked in and reduced the gain (per the service manual). For load testing, I stopped turning the volume knob as soon as the "0dB" light came on and didn't push it into clipping. Now for some cheap thermal camera fun.. For some reason the browser/ forum keeps showing the photos rotated vs. what's on my PC. I've rotated them several times to try and fix the formatting, but they are always the wrong orientation! This was very shortly after the load test- inside the chassis is glowing well, but the power resistors were really putting out some heat! 2
Half Life 256 Posted March 26 Posted March 26 On 3/24/2025 at 5:45 PM, MattB said: .. unfortunately, the amp is idling a bit hot- it's drawing ~1.1A at idle and a fair amount of heat is originating from the center of the chassis looks like Q5 and Q6 on the right channel when I sweep my thermal camera across the chassis. A thermal camera is a good tool when troubleshooting. I have a FLIR C2 and use it quite a bit when repairing various pcb's. It helped me find a diode that was breaking down in a TFM-15 I repaired several years ago. Nice work! Enjoyed the thread! 1 2
MattB 34 Posted March 27 Author Posted March 27 17 hours ago, Half Life said: A thermal camera is a good tool when troubleshooting. I have a FLIR C2 and use it quite a bit when repairing various pcb's. It helped me find a diode that was breaking down in a TFM-15 I repaired several years ago. Nice work! Enjoyed the thread! Thanks! I hoped my escapades would maybe help someone else and provide some nifty photos that not all people are willing to do.. I have no problems ripping into a project and doing a complete teardown that most people just say "send it off to a pro". I'm too impatient for that and would rather spend two weekends screwing it up myself instead of just paying a couple hundred bucks to a professional. This little thermal camera is very handy- troubleshooting is pretty much why I bought this thing. it's not a very high end camera- just one of those little $100 strap-ons for my phone, but it's great to search for faults in electronics or diagnosing a misfire on the race car (just look for which header tube is different than all the others!) I also diagnosed some bad panels on a solar array. Nifty lesson there- warm spots on a solar panel are where they aren't producing energy! The panels are cooler because they're converting the solar energy to electrical and shipping it down a wire. the places there the panel isn't converting light into electricity, it converts the light to heat and is thus warmer. It's a great cheat playing hide and seek with the kids since they leave tiny warm footprints in the carpet for a couple minutes! 2 2
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now